Prenatal Sensitization in a Mouse Model

UDO HERZ, RICARDA JOACHIM, BIRGIT AHRENS, ALEXANDER SCHEFFOLD, ANDREAS RADBRUCH, and HARALD RENZ

Department of Clinical Chemistry and Molecular Diagnostics, Clinic of the Philipps-University Marburg, Marburg, Germany; and Deutsches Rheumaforschungszentrum, Berlin, Germany

	INTRODUCTION

TOP INTRODUCTION RESULTS AND DISCUSSION CONCLUSION DISCUSSION REFERENCES

Several studies have suggested that the in utero environment plays an important role in the onset of sensitization to environmental allergens. This concept of in utero priming to favor the development of subsequent allergic diseases is supported by several observations. Already at birth cord blood mononuclear cells (CBMCs) proliferate in response to allergens such as Dermatophagoides pteronyssinus (Der p) or birch pollen (1). This allergen-specific immune response is mediated by fetal T cells, as opposed to maternally derived contaminants (4). Several factors remained crucial for the development of such a proliferative immune response. The mother must be exposed to the antigen beyond the first trimester of pregnancy (5, 6). The presence of allergen-specific IgE at birth also indicates the prenatal development of functional T-B cell interactions (7). This implies that the allergen responder phenotype may be determined by immune deviation involving allergen-driven T cell selection. Such a notion is supported by finding that pregnancy is associated with a transient depression of maternal helper T cell type 1 (Th1)-mediated cell immunity, resulting in a Th2-like maternal profile at the maternal-fetal interface. Moreover, many studies have documented defective polyclonal interferon  (IFN-) responses by CBMCs from neonates either at risk of atopy, or those who later develop disease (8).

Our aim was to analyze the development of antigen-specific T and B cell responses in the pre- and postnatal immune system in a defined animal model of allergic sensitization. In this BALB/c mouse model, sensitization to ovalbumin (OVA) induces allergen-specific IgE/IgG₁ production associated with allergic response in the airway and increased airway hyperreactivity. This model was used to sensitize female BALB/c mice to OVA before their mating with nonsensitized BALB/c males. During pregnancy, mice were exposed to allergen aerosol every second day. Offspring T and B cell responses were analyzed at various time points after birth.

	RESULTS AND DISCUSSION

TOP INTRODUCTION RESULTS AND DISCUSSION CONCLUSION DISCUSSION REFERENCES

After OVA sensitization, anti-OVA IgE, IgG₁, and, to a lesser degree IgG_2a serum antibody titers were induced (p  0.05). Because of the presence of allergen-specific IgE/IgG₁ antibodies, pregnant BALB/c mice developed immediate-type cutaneous hypersensitivity (ICHS) responses to OVA (12).

To determine whether this Th2 type-driven immune response is associated with an imbalance in the Th1/Th2 response, the frequencies of interleukin 4 (IL-4) and/or IFN--producing T cells were analyzed at the single-cell level by intracellular cytokine staining. A protocol of 6-h stimulation with phorbol myristate acetate (PMA) plus ionomycin was selected because this protocol resulted in recall of memory T cells (13). No significant difference in the frequencies of IFN-- and IL-4-producing T cells was detected between allergen-sensitized and nonsensitized mice. In both groups, 25.8% and 27.6% of all T cells produced IFN- and about 1-3% produced IL-4 (12). Therefore, on the level of memory T cells, no imbalance in the Th2/Th1 cytokine profile could be detected at the time of delivery. To analyze the development of immune responses in offspring, spleens were obtained and the numbers of T and B cells were monitored up to 60 days postpartum. T cell ontogenesis markedly differs between human and mouse. In the human immune system T cell development takes place during the first trimester. At the end of the first trimester a small number of mature CD4- or CD8-expressing T cells can be detected. The second and third trimesters are characterized by colonization of peripheral lymphoid tissues with mature T and B cells and expansion of the peripheral lymphocyte pool. In contrast, in mice the first set of mature CD4- or CD8-expressing T cells appears just 1 or 2 days before birth (Figure 1 [14]). Therefore, mice are born with a small number of mature lymphocytes, and expansion of the peripheral lymphocyte pool occurs in the postnatal period. An about 100-fold expansion of T cells was detected during the first week of life. The rapid expansion of the peripheral lymphocyte pool is associated with a high rate of lymphocyte proliferation as detected by [³H]thymidine incorporation. This situation makes it difficult to assess an antigen-specific T cell response at birth because of the low frequency of mature T cells and the high proliferation rate of the rapidly expanding lymphocyte population. However, assessment of OVA-specific lymphocyte proliferation rates indicated the presence of OVA-specific T cells already at birth.

View larger version (20K): [in this window] [in a new window]   Figure 1.   T and B cell ontogenesis in humans and mice. (Adapted from Paul, W. E., editor. Fundamental Immunology, 3rd edition. Raven Press, New York [14].)

The cytokine profile (IL-4, IFN-) in splenic mononuclear cells (MNCs) from offspring was analyzed at the single cell level by intracellular staining on days 1-2 postpartum, as described above. In offspring from nonsensitized BALB/c mice, 4.1% of the Thy1⁺ cells could be restimulated to produce IFN-. In contrast, offspring from OVA-sensitized BALB/c mice showed a fivefold reduced capacity for IFN- production (0.8%). However, no significant difference in the frequency of IL-4-producing T cells was detectable (< 1%) (12). These data demonstrate the presence of IFN--producing effector T cells already at birth. IFN- production was not the result of in vitro differentiation of T cells because we used a protocol that recalled cytokine production in already differentiated effector T cells. Compared with adult mice the frequency of IFN--secreting T cells is significantly reduced at birth. It is unlikely that this is due to a delayed maturation of T cells because only Thy1-positive T cells were analyzed. It is more likely, however, that the reduced frequency of IFN--secreting effector T cells is the result of regulatory events that prevent the development of strong Th1 immune responses during pre- and postnatal periods. Another important finding from our study is that mice from OVA-sensitized mothers demonstrate an even more pronounced reduction in IFN--secreting effector T cells than mice from nonallergic mothers.

In parallel with T cell functions, we also analyzed the development of B cell responses. To compare the transfer of immunoglobulin isotypes (IgE/IgG₁/IgG_2a) in mice and humans, serum antibody titers in offspring from OVA-sensitized BALB/c mice were analyzed. Three patterns of active diplacental/transintestinal transfer were observed during pre- and postnatal maturation: (1) allergen-specific IgE antibodies could not be detected in offspring at any time; (2) combined diplacental plus postnatal transfer was found for allergen-specific IgG₁ antibodies; and (3) exclusive prenatal diplacental transfer was detected for allergen-specific IgG_2a antibodies. The postnatal transfer is most likely explained by antibody uptake from breast milk and transport via the FcRn receptor. The FcRn receptor mediates transfer of maternal, milk-borne IgGs across the rodent neonatal intestine (15). In addition, these offspring responded with positive immediate-type skin test reactions to OVA. On the basis of the preceding findings, these positive immediate-type skin test responses in offspring are most likely triggered by the presence of maternal OVA-specific IgG₁ antibodies. Also, maternal IgG₁ transfer simultaneous to allergen exposure during pregnancy may support priming of the allergic phenotype iin offspring. Possible mechanisms may include the formation of immune complexes (Figure 2).

View larger version (34K): [in this window] [in a new window]   Figure 2.   A concept of pre- and postnatal regulation of Th1 and Th2 immunity.

	CONCLUSION

TOP INTRODUCTION RESULTS AND DISCUSSION CONCLUSION DISCUSSION REFERENCES

We have developed a murine animal model system that allows analysis of the impact of maternal allergy on the development of physiological and pathological immune responses in offspring. Analysis of the physiological ontogenesis of T and B cell functions in mice revealed that T cell development is completed just before birth and a few mature CD4- or CD8-expressing T cells appear in the periphery at the time of birth. The low frequency of these T cells is already primed to produce IFN-, an important Th1 cytokine. This reduced frequency of IFN--secreting T cells within the pool of mature lymphocyte is most likely the result of regulatory events that suppress Th1 immunity in the pre- and postnatal period. Because IFN- is not only an important Th1 cytokine with para- and autocrine functions but also one of the most potent inhibitors of Th2 immunity, the reduced frequency of IFN--secreting T cells results in a lack of inhibition of Th2-type responses. Therefore, the missing negative signal favors the development of Th2 immunity in the presence of appropriate antigen-specific and costimulatory signals. In offspring from allergic mothers the Th1 capacity is reduced even further compared with mice from normal mothers. We propose that this reduced IFN- capacity is a key factor in the development of atopy. It is important to note that the Th1 effector T cells present at birth are not necessarily allergen-specific T cells. This suggests, on the one hand, that the presence or absence of these Th1 cells plays a more general role in the regulation and maturation of effector T cells in the postnatal period; and, on the other hand, this finding raises the question of how this phenomenon is regulated. A wide range of mechanisms must be considered, including antibody transfer of certain isotypes, transfer of immune complexes, and transfer of cytokines and other soluble molecules via the placental membrane (Figure 2).

	DISCUSSION

TOP INTRODUCTION RESULTS AND DISCUSSION CONCLUSION DISCUSSION REFERENCES

Holt: We looked at the model in a much more cursory way than you have and can confirm your findings. To get these signals we, like you, had to use two or three cycles of priming with Al(OH)₃, which leaves a large depot of allergen. It is a very big stimulus to these animals. You an actually measure circulating IL-4 in these animals, and IL-5 at sometimes quite high levels. It is not unrealistic that some of this might cross the placenta. We have some evidence that this can occur. Might not this system be an exaggeration of what might occur in humans?

Renz: One has to be careful, I agree. What we tried to do is, first of all, to develop an optimized system and then we will go back and see what are the minimal requirements. But what I would like to emphasize is: we primed the mothers before they were pregnant. This was done i.p. with Al(OH)₃, but during pregnancy they were just aerosol-exposed every other day to mimic the natural route of exposure to aerollergens. So, we were very careful to avoid adjuvants at this time. I think this is all that is required: allergen through the airways. We don't yet know about the timing: is it the first or second period during pregnancy? We believe it is the second period, but is this enough to give us the same type of response? But, again, I agree we have to be careful about using adjuvants.

Björkstén: In our collaborative study between Perth and Linkoping, Dr. Jenmalm is looking at IgE and IgG subclass antibody responses to inhalants and to food over the first 8 years of life, in relation to atopy. What she has found is that high levels of IgG subclasses in the cord blood are related to less allergy and a nice correlation between cord blood levels for inhalants and allergy at 8 years. This was true for birch and cat, the inhalants, but it was not true for the foods. A possible explanation is that there is a downregulatory mechanism for the foods.

Renz: I agree. The respiratory mucosa might be different from the intestinal mucosa. We did a small study, together with Dr. Wahn, where we analyzed cord blood T cell responses. We found T cell responses in normal neonates. We could find these T cell responses only to food antigens, but not to inhalant antigens, although exposure was there. Whether this is related to the dose, to the timing, or to immunlogical reasons, intrinsic reasons, we do not know yet.

Holgate: Your animal model provides a unique opportunity to investigate factors other than maternal allergen sensitization and exposure on the evolution of allergic disease in the offspring. Such factors include maternal diet (e.g., over- rather than under-nutrition), exposure to tobacco smoke. Are you planning to expand your model into these areas?

Renz: Yes, this viewpoint actually justifies why we need animal models: in order to control these different factors that contribute. The genetics are very controlled. These studies have been done in genetically high-responder animals. Now the question is: How important is the genetic imprinting of this phenomenon; can we achieve the same phenotype in genetically not at risk populations?

Djukanovic: The T cell responses that you see in the fetus must represent clonal expansion of T cells that are specific for that antigen. How does the T cell of the fetus see the antigen. It does make sense that the mother prepares the offspring for the environment, but how it confers this specific immunity is puzzling. We like to think about antigen-presenting cells as cells that do their thing locally and perhaps are talking to the draining lymph nodes, but is there a way whereby antigen can be picked up by the mother and circulate throughout the body, pass then through the placenta?

Renz: There is little or no evidence that antigen is really circulating in the fetus, but there is quite good evidence that immune complexes are transported across the placenta, and can actually be taken up by these FcRn receptors, that are not only expressed on the intestinal surfaces of the neonate but also on the placenta itself. So, at least in mice, the placenta carries a receptor that specifically can take up immune complexes and transport them through his barrier. This is one mechanism.

Djukanovic: This is with ovalbumin, but what about house dust mite?

Renz: If it is possible for these kind of antigens, I would assume that it can also take place for the house dust mite. But it has not been investigated in the human.

Björkstén: Dr. Casas published last year on Fel d 1 immune complexes in asthmatic children and now she submitted a manuscript describing Fel d 1 immune complexes in the cord blood. The interesting thing is that there is no difference that we can see between atopic and nonatopic mothers, so it is a fairly common finding, but we cannot relate it to atopy, in the mother at least.

Renz: Exactly. This takes place, but we don't know whether it is an epiphenomenon or causally related to the development of a specific immune response.

Holt: There is no doubt now that this can happen under appropriate modes of antigen presentation. The vaccine industry is taking this quite seriously, in view of demonstrations that you can get IgM responses to tetanus in the fetus as the result of immunization of the mother. Of course, that is not the same as allergen exposure, but the principle is there.

Renz: This may not be the only mechanism, of course. Soluble factors like cytokines can also pass this barrier. There are some hints that even antigen-presenting cells can also travel. I think this is an open field.

Knol: If you adapt your model to drive the mother in the prenatal period more towards the Th1 response, would you then also be able to help the offspring, will it develop less allergy?

Renz: That would be the hypothesis, but we have not tested it yet.

Platts-Mills: If this is going on in reality, with inhalant allergens, then by any mechanism the state of immunity in the mother has got to influence what happens to the baby. You could either have antigen transfer without immune complexes or with immune complexes, but my guess is that with immune complexes it would be much more effective. The proposal would be that the mother would influence the development of atopy in the infant more than the father. So the question is: Is maternal influence over asthma in the child affecting allergic children or nonallergic children? I would turn that question round to Dr. Martinez: Is the excessive effect of maternal immunity at age 6 in the Tucson cohort on the allergic children or on the nonallergic children?

Martinez: This is a very important point. Levels of IgE in the child strongly correlate with those of the mother at birth, at age 6 and at age 13, but in a descending strength. The lowest, at age 13, is the same level of correlation that the father shows with IgE levels in the child. That is the first piece of information. The second piece of information is very important and is extremely politically incorrect, unfortunately. Mothers who are allergic and breast feed have more allergic children later than mothers who are allergic and do not breast feed. Mothers who are allergic and breast feed have children with higher levels of IgE and who are more likely to be sensitized at the age of 6 than those who aren't. The interesting thing is that when you go to age 13, this is not the case, so it appears that what the mothers are doing, by some in utero influence or through the breast milk, is favoring earlier allergic sensitization. This takes me back to what I have been saying before with respect to the timing of sensitization.

Renz: The mouse data indicate that maternal imprinting takes place in terms of shaping the immune system of the neonate, both phenotypically and functionally. This effect diminishes after birth. There is a window of opportunity shortly after birth.

Platts-Mills: One more comment about the data of Dr. Martinez. In children with Alternaria sensitization and asthma the phenotype of both parents seems to have an equal effect. By contrast, the mother has more influence on wheezing in nonallergic children. Is that correct?

Martinez: That is precisely what we have been discussing before. This refers to asthma. There is a specific influence on the way that the lung develops.

Platts-Mills: But this is the really important thing. If we are going to see a maternal influence, and it is really working through this kind of mechanism, it has got to work on atopy and that has not been shown.

Savelkoul: There are some references in the literature that there is not only an isotype-specific transfer over the placenta, but even perhaps an idiotypic-specific transfer. With tetanus-specific antibodies or the anti-polysaccharide antibodies you find in the offspring only the highest affinity antibodies from the mother. Did you compare the IgG₁ isotype spectrum of antibodies present in the mother and in the pups after birth?

Renz: It becomes now clear that there is very little data on what is actually going on in this critical interface. All the good data that is available indicates selectivity.

Weiss: There could be a reciprocal relationship: an effect of the fetus on the maternal immune system. A certain percentage of women who have never been allergic (and I want to make it clear that I am saying allergic and not asthmatic) develop allergy often during the second or third trimester of pregnancy. Conceivably, this could be related to fetal transfer of cells. There are other examples in the immunologic literature. There is the thought that some autoimmune diseases, scleroderma, may be related to retention of fetal cells and the development of autoimmunity.

Renz: There is definitely a suppression of Th1 immunity and DTH responses during pregnancy, particularly during the second and third trimesters. It is apparently related to the high levels of progesterone and the production of -fetoprotein, but it might also be directly influenced by fetal immunity.

	Footnotes

Correspondence and requests for reprints should be addressed to H. Renz, M.D., Clinical Chemistry and Molecular Diagnostics, Central Laboratory, Hospital of the University, Baldingerstrasse, 35033 Marburg, Germany. E-mail: renzh{at}post.med.uni-marburg.de

	References

TOP INTRODUCTION RESULTS AND DISCUSSION CONCLUSION DISCUSSION REFERENCES

1. Kondo, N., Y. Kobayashi, S. Shinoda, K. Kasahara, T. Kameyama, S. Iwasa, and T. Orii. 1992. Cord blood lymphocyte responses to food antigens for the prediction of allergic disorders. Arch. Dis. Child. 67: 1003-1007 [Abstract].

2. Piccinni, M. P., F. Mecacci, S. Sampognaro, R. Manetti, P. Parronchi, E. Maggi, and S. Romagnani. 1993. Aeroallergen sensitization can occur during fetal life. Int. Arch. Allergy Immunol. 102: 301-303 [Medline].

3. Szepfalusi, Z., I. Netwich, M. Gerstmayr, E. Jost, L. Todoran, R. Gratzl, K. Herkner, and R. Urbanek. 1997. Prenatal allergen contact with milk proteins [see comments]. Clin. Exp. Allergy 27: 28-35 [Medline].

4. Prescott, S. L., C. Macaubas, B. J. Holt, T. B. Smallacombe, R. Loh, P. D. Sly, and P. G. Holt. 1998. Transplacental priming of the human immune system to environmental allergens: universal skewing of initial T cell responses toward the Th2 cytokine profile. J. Immunol. 160: 4730-4737 [Abstract/Free Full Text].

5. Jones, A. C., E. A. Miles, J. O. Warner, B. M. Colwell, T. N. Bryant, and J. A. Warner. 1996. Fetal peripheral blood mononuclear cell proliferative responses to mitogenic and allergenic stimuli during gestation. Pediatr. Allergy Immunol. 7: 109-116 [Medline].

6. Van Duren Schmidt, K., J. Pichler, C. Ebner, P. Bartmann, E. Forster, R. Urbanek, and Z. Szepfalusi. 1997. Prenatal contact with inhalant allergens. Pediatr. Res. 41: 128-131 [Medline].

7. Bergmann, R. L., G. Edenharter, K. E. Bergmann, I. Guggenmoos, Holzmann, J. Forster, C. P. Bauer, V. Wahn, F. Zepp, and U. Wahn. 1997. Predictability of early atopy by cord blood-IgE and parental history. Clin. Exp. Allergy 27: 752-760 [Medline].

8. Rinas, U., G. Horneff, and V. Wahn. 1993. Interferon-gamma production by cord-blood mononuclear cells is reduced in newborns with a family history of atopic disease and is independent from cord blood IgE-levels. Pediatr. Allergy Immunol. 4: 60-64 [Medline].

9. Tang, M. L., A. S. Kemp, J. Thorburn, and D. J. Hill. 1994. Reduced interferon-gamma secretion in neonates and subsequent atopy [see comments]. Lancet 344: 983-985 [Medline].

10. Martinez, F. D., D. A. Stern, A. L. Wright, C. J. Holberg, L. M. Taussig, and M. Halonen. 1995. Association of interleukin-2 and interferon-gamma production by blood mononuclear cells in infancy with parental allergy skin tests and with subsequent development of atopy. J. Allergy Clin. Immunol. 96: 652-660 [Medline].

11. Liao, S. Y., T. N. Liao, B. L. Chiang, M. S. Huang, C. C. Chen, C. C. Chou, and K. H. Hsieh. 1996. Decreased production of IFN and increased production of IL-6 by cord blood mononuclear cells of newborns with a high risk of allergy. Clin. Exp. Allergy 26: 397-405 [Medline].

12. Herz, U., B. Ahrens, A. Scheffold, R. Joachim, A. Radbruch, and H. Renz. 2000. Impact of in utero Th2 immunity on T cell deviation and subsequent immediate-type hypersensitivity in the neonate. Eur. J. Immunol. 30: 714-718 [Medline].

13. Assenmacher, M., J. Schmitz, and A. Radbruch. 1994. Flow cytometric determination of cytokines in activated murine T helper lymphocytes: expression of interleukin-10 in interferon-gamma and in interleukin-4-expressing cells. Eur. J. Immunol. 24: 1097-1101 [Medline].

14. Paul, W. E., editor. 1993. Fundamental Immunology, 3rd ed. Raven Press, New York.

15. Jones, E. A., and T. A. Waldmann. 1972. The mechanism of intestinal uptake and transcellular transport of IgG in the neonatal rat. J. Clin. Invest. 51: 2916-2927 .